Leon R. Roose, Esq.

Principal & Chief Technologist, GridSTART

Hawaii Natural Energy InstituteSchool of Ocean & Earth Science & Technology

University of Hawaii at Manoa1680 East-West Road, POST 109

Honolulu, Hawaii 96822

Utility Scale Storage ApplicationsThe Hawaii Experience

Utility-Scale Storage: If, When, What Type, How Much, and Where?

Asia Clean Energy ForumJune 5, 2018

Manila, Philippines1

Hawaii Natural Energy Institute (HNEI)University of Hawai‘i at Mānoa

Organized Research Unit in School of Ocean and Earth Science and Technology

Founded in 1974, established in Hawai‘i statute in 2007 (HRS304A-1891)

• Conduct RDT&E to accelerate and facilitate the use of resilient

alternative energy technologies; and to reduce Hawaii’s dependence on

fossil fuels.

• Diverse staff includes engineers, scientists, lawyers; students and

postdoctoral fellows; visiting scholars

Areas of Interest

• Policy and Innovation

• Grid Integration (GridSTART)

• Alternative Fuels

• Electrochemical Power

Systems

• Renewable Power Generation

• Building Efficiency

• Transportation

Core Functions

• State Energy Policy Support

• Research & Development

• Testing and Evaluation

• Analysis

• Workforce Development

2

• Serves to integrate into the operating power grid other HNEI

technology areas: biomass and biofuels, fuel cells and hydrogen,

energy efficiency, renewable power generation

• Strong and growing partnerships with national and international

organizations including Asia-Pacific nations.

Established to develop and test advanced grid architectures, new

technologies and methods for effective integration of renewable

energy resources, power system optimization and enabling policies.

Lead for many public-private demonstration projects

3

Hawaii’s Progressive Leadership in

Clean Energy Policy

Highest RPS Target

in the United States

100% by 2045(2015 - 15%; 2020 - 30%, 2030 – 40%, 2040 – 70%)

Other key policies:

• Tax incentives

• Net metering

• Feed in tariffs

• Decoupling

Strong Hawaii Policies

Hawaii Clean Energy Initiative (HCEI)The State of Hawaii, US DOE, and local utility launched

HCEI in January 2008 to transform Hawaii to a 70%

clean energy economy by 2030:

• Increasing Hawaii’s economic and energy security

• Fostering and demonstrating Hawaii’s innovation

• Developing Hawaii’s workforce of the future

• Becoming a clean energy model for the U.S. and the

world

4

Oʻahu

Molokaʻi

Maui

HawaiʻiLanaʻi37%

19%

41%

54%

% Renewable

Energy

Kauaʻi Island Utility CooperativeSystem Peak: 78 MW

65.6 MW PV / 7 MW Biomass / 9 MW Hydro

Installed PV: 84% of System Peak

41.7% RE in 2016

Maui ElectricMaui System Peak: 202 MW

102 MW PV / 72 MW Wind

Installed PV & Wind:

86% of Sys. Peak

34.2% RE in 2017

Lana’i System Peak: 5.1 MW

2.53 MW PV (50% of Sys. Peak)

Moloka’i System Peak: 5.6 MW

2.3 MW PV (41% of Sys. Peak)

Hawaiian ElectricSystem Peak: 1,206 MW

512 MW PV / 99 MW Wind / 69 MW WTE

Installed PV & Wind:

50% of System Peak

20.8% RE in 2017

Hawaii Electric LightSystem Peak: 192 MW

92 MW PV / 30 MW Wind /

38 MW Geothermal / 16 MW Hydro

Installed PV & Wind:

64% of System Peak

56.6% RE in 2017

Kaua’i

80% of state population

Hawai‘i Electric Systems –4 Electric Utilities; 6 Separate Grids; % Renewable Energy

5

Hawaii Energy Storage Projects

Utility project

IPP or Third-party project

Kaheawa I BESS1.5 MW/1 MWhXtreme PowerAdvanced lead acid

Kaheawa II BESS10 MW/20 MWhXtreme PowerAdvanced lead acid

MECO Smart Grid BESS1 MW/1 MWhA123Lithium Ion

HELCO Power Conditioner2 x 100 kW/248 kWhSaftLithium ion

Auwahi Wind BESS11 MW/4.4 MWhA123Lithium Ion

HNEI/HELCO Hawi BESS1 MW/250 kWhAltairnanoLithium titanate

La Ola PV BESS1.125 MW/0.5 MWhXtreme PowerAdvanced lead acid

HNEI/MECOMolokai BESS2 MW/397 kWhAltairnanoLithium titanate

HECO PV/BESS EV Carport6 kW/20 kWhThunderSky/GreensmithLithium ion

HNEI/HECOCIP BESS Demo1 MW/250 kWhAltairnanoLithium titanate

Koloa BESS1.5 MW/1 MWhXtreme PowerAdvanced lead acid

Port Allen BESS3 MW/2 MWhXtreme PowerAdvanced lead acid Anahola BESS

6 MW/4.6MWhSAFT Li-ion

HNEI projects

Hawaii CAISO

Peak (MW) 1,578 31,144

Storage (MW) 53 1,300

% Storage 3.4% 4.2%

Solar City/Tesla BESS13 MW/52MWhSAFT Li-ion

6

Maui IslandLeading the way in Wind and Solar Power

Kaheawa I

(30 MW)

Kaheawa II

(21 MW)

Auwahi

(21 MW)

102 MW Existing Distributed PV

~12 MW PV Pre-approved

114 MW Total

Wind - 72 MW

PV - 114 MW

186 MW

7

Mitigation Measures to Increase Wind Energy Delivered on Maui

BESS Function

• 10MW / 20MWh

• Manual and AGC Dispatch

• Aggressive frequency Response

• Ramp Rate Limit within a limited SOC

Range

MECO Operations

• Include 10MW of BESS in Up Reserve

• Reduce Down Reserve of M14 & M16 by 1.5MW

• Reduced Operation of K1 and K2

• 50MW Up-Reserve Limit

FacilityPercent of available energy delivered

(% before / % after)

% increase

in delivered

energy

Plant 1 97% / 99% 2%

Plant 2 72% / 84% 17%

Plant 3 27% / 45% 68%

8

Higher Power/Limited Energy BESS application to meet grid upward reserve needs was most cost effective

HNEI/HEI (Hawaii electric utility) BESS Projects

CIP BESS

Moloka’i BESS

Hawi BESS

9

Hawi, Hawaii Island (1MW, 250kWh)

• Modeling showing benefit completed in 2007

• Frequency regulation and wind smoothing

• Rapid and extended cycling

Molokai Secure Renewable Microgrid (2MW, 375kWh)

• Operating reserves (fault management), frequency regulation,

• Fast response decision and control (<50ms response)

Campbell Park industrial feeder with high penetration (1MW)

• Power smoothing, voltage and VArsupport, and frequency regulation

Demonstrate BESS operating strategies for high value grid applications

Hawaii Island BESS Project

• Island-wide power flow modeling indicated that a 1 MW fast-acting battery could significantly reduce grid frequency variability

• HNEI engaged public and private entities to develop a 1MW BESS on Hawaii Island at the interface between a 10 MW wind farm and the grid

• Commissioned December 2012 with testing initiated in spring 2013

Photo courtesy of Altairnano, Inc.

60.00

60.05

59.95

59.90

59.85

Grid F

requency [

Hz]

0 800 1600 2400

Time [Seconds]

10

Summary Field Results (Hawaii Island BESS)

6:00 PM HST 4/9/2015 to 9:20 PM HST 4/9/2015. Max

Power = 1000kW, No dead band, Gain = 30MW/Hz

Power smoothing from 10MW Hawi wind farm

Frequency regulation: Up to 40% reduction in f fluctuation

Battery response ~ 250ms

11

Renewable Moloka’i Initiative

• 100% Renewable Goal by 2020

• Battery Storage

• 2MW, 333kW-hr, Li-ion Titanate

• System Data Collection

• Production Modeling

• Load Flow & Midterm Dynamics Modeling

• Dynamic Load Bank

• PV Forecasting

• Island Grid Controller

12

• Peak Load: 5.4 MW (2013)

• Minimum Load: 3 MW

(daytime, 2009)

• PV Installed: 1.07 MW (with

59.3Hz drop-out)

• PV Installed: 1.23 MW (with

57Hz drop-out)

• PV Planned: 0.6 MW in que

Molokai Grid Stability• Even relatively small disturbances can trip PV units, increasing automatic load

shedding customer outages on the system.

• Proposed solution to increase grid reliability: a 2MW fast-acting BESS.

• The challenge: standard 250 ms response destabilizing to grid (models)

• The solution: re-engineer the way the BESS and the inverter computers collaborate to share computational burden

(Inverter)

(Battery)

13

• Power module produces ± 2 MW

• Capacity of 375 kW-Hr

• Inverter rated > ± 2 MVA

• Over 12,000 full charge / discharge

cycles with minimal degradation in

cell capacity

Interior View

• Li-Ion Titanate 50 A-Hr Cells

• BESS has 2688 cells in 96

LRU

Power Module Power Conversion System

Molokai Island BESS ProjectAltairnano Li-ion Titanate

Designed for rapid charging and discharging

14

Impact of BESS with 500kW Authority and Fast ResponseExpansions (bottom) show Down-Step

BESS Authority = 500kW

BESS Authority = 500kW

BESS Off

500kW fast response showed significant frequency improvement15

Intermittent Grid Frequency Oscillation• A grid anomaly of unknown source has

occurred 7 times since September 2016.

• In one instance, on May 12th, 2017, the BESS was online with a limit of 500kW.

• Because the oscillation rate was about twice as fast as the response time of the BESS, the oscillations were exacerbated by the BESS.

• Molokai grid operators set the BESS offline around 87 seconds after the oscillations started.

• An oscillation detection subroutine was developed to suppress BESS responses at the onset of oscillations

16

The real world consequences of the BESS running with a higher

authority needs to be better understood.

Observations Based on HI Experience• Energy storage is and will be key to meeting RE goals

‒ Variety of applications to integrate high levels of variable renewables

• Still a relatively early commercial technology

‒ Technical and market risks remain (Kahuku fire and vendor turnover)

‒ Still difficult to get standardized pricing

‒ Technical performance not always as advertised

‒ Battery lifetime/degradation still uncertain

• Rapid industrial scaling of industry underway and will address many of the early stage risks and uncertainties

17

HNEI/GE BESS Analyses for Oahu

25 50 100 200

0.50 12.5 25 50 100

1.00 25 50 100 200

2.00 50 100 200 400

4.00 100 200 400 800

Power (MW)Energy (MWh)

Stora

ge (H

rs)

Power and Energy Rating of BESS Considered

Objective: Quantify net benefits of various BESS configurations on Oahu System

SCEN 1 (Current W&S)

SCEN 2(50% W&S)

Fuel Price ($/bbl) $60/bbl WTI = ~$11/MMBtu

Thermal Resources Existing Units, No Adds or Retires

Peak Load (MW) 1,225 1,225

Annual Energy (GWh) 7,734 8,450

Electric Vehicles (GWh) 44 791

Wind & Solar Capacity (MW) 809 1965

Utility-Scale Wind 123 565

Utility-Scale Solar 148 565

Distributed PV 538 840

Available W&S (GWh) 1547 4225

Available W&S (% of Load) 20% 50%

Assumptions for Two Scenarios Analyzed

18

Production Cost Savings vs Annual Capital Cost by Configuration

• Larger Circle – more MWh• Farther from Line – higher

benefit-to-cost ratio• High Power/Low Energy

Configurations most cost effective

19

Key Findings – Oahu BESS Analysis Operation and control of storage can be co-optimized to provide both

reserves and energy arbitrage

Charging mostly takes place during mid-day hours, discharge during evening peak, but also some charging during overnight periods and discharging during morning ramp

Energy storage will increase generation from wind and solar (decreased curtailment) along with more efficient operation of thermal resources (Kalaeloa CC, AES)

Increasing power rating (inverters) is less expensive than increasing energy rating (lithium-ion cells)

Economic cost-benefit analysis favors high-power, low-energy ratio storage configurations

Energy storage can be economic in some (but not all) configurations with current W&S scenario, and all configurations for 50% wind and solar

While this analysis shows that BESS can be an economic addition to Oahu’s resource mix, it is not necessarily the most economic, other options to analyze and compare: Demand Response, Increased flexibility in thermal fleet, etc.

20

Alternatives to Storage

21

Small to moderate amount of excess RE curtailment is a sound integration strategy

Mahalo!(Thank you)

For more information, contact:

Leon R. Roose, Esq.

Principal & Chief TechnologistGridSTART

Hawaii Natural Energy InstituteSchool of Ocean & Earth Science & TechnologyUniversity of Hawaii at Manoa1680 East-West Road, POST 109Honolulu, Hawaii 96822

Office: (808) 956-2331Mobile: (808) 554-9891E-mail: lroose@hawaii.eduWebsite: www.hnei.hawaii.edu

Grid System Technologies Advanced Research Team

22